# 2009 State of the Market Report

Transmission Congestion

Due to the physical properties of electricity, more power tends to flow over higher-voltage lines, and a wider array of generators tends to affect these flows. Conversely, low voltage constraints typically must be managed with a smaller set of localized generating resources, making low- voltage constraints more difficult to manage. Figure 58 shows constraint manageability improved substantially at all voltages equal to or greater than 138 kV, which constitutes the vast majority of the congestion on the Midwest ISO system. These improvements are attributable to the changes discussed above.

Only the lowest voltage level exhibited worse manageability in 2009. Nearly 40 percent of low- voltage facilities (those rated 69 kV to 115 kV) were unmanageable in 2009, up from 33 percent in 2008). This suggests that the Midwest ISO accepted responsibility for facilities that it lacked the resources to effectively manage. We recommend that the Midwest ISO establish criteria for determining when it should secure these low voltage facilities and when they are more appropriately secured by local balancing authorities.

Given the frequency with which constraints are unmanageable, it is very important that the congestion be priced efficiently in the Midwest ISO’s LMPs. Before we evaluate this, it is important to understand how the real-time market treats transmission constraints. The real-time market model utilizes “marginal value limits” (“MVL”) that represent the value of managing the constraint. The MVL caps the cost (i.e., the shadow price) that the real-time market will incur to reduce the constraint flow to the limit. Hence, it is the maximum cost the Midwest ISO would be willing to incur to manage the constraint. Presumably, therefore, it should reflect the true reliability cost of violating the constraint. When the constraint is violated, the most efficient LMPs would be those that reflect the MVL of the violated constraint.

Figure 59 evaluates the pricing of violated constraints be showing how consistent shadow prices of the constraints have been with the MVLs. In this figure, the violated constraint hours are divided into tranches by the ratio of the shadow price to the MVL of the constraint. The ratio of the constraint shadow price to the MVL determines the extent to which the shadow price fully reflects the cost of the violated constraint. When the shadow price is close to 100 percent of the MVL, it accurately reflects the congestion on the violated constraint. When the ratio is significantly less than 100 percent, the congestion indicated in the LMPs is inefficiently muted.

Page 89